Analytics

Orbit’s Analytics endpoints expose the same delivery, engagement, and cost datasets that power the dashboard’s Insights page. Use them to embed Orbit metrics in your internal BI tooling, schedule exports to a data warehouse, or programmatically alert on delivery / spend regressions. All analytics endpoints live under /api/v1/analytics, require an API key with the analytics:read scope, and are aggregated against a read replica so heavy reporting queries never contend with the live send path. Each endpoint is rate-limited at 60 requests / minute per tenant. For the model behind the surface — what the aggregates expose, why the read-replica separation holds, how the background roll-up producers pre-aggregate the series, and how the dashboard Insights pages consume the same datasets — see the analytics pipeline concept. Every response uses the standard Orbit envelope: the payload is under data, and meta carries the request_id + timestamp.

Query delivery metrics

Aggregate sent / delivered / failed / read counts with a time-series breakdown. Filter by channel, status, or campaign_id, and bucket the series with group_by (hour, day, week, month — defaults to day).

curl -G https://api.orbit.devotel.io/api/v1/analytics/messages \
  -H "X-API-Key: dv_live_sk_..." \
  --data-urlencode "channel=sms" \
  --data-urlencode "status=delivered" \
  --data-urlencode "start_date=2026-05-01T00:00:00Z" \
  --data-urlencode "end_date=2026-05-08T00:00:00Z" \
  --data-urlencode "group_by=day"

Response

{
  "data": {
    "totals": {
      "total_sent": 39534,
      "total_delivered": 38901,
      "total_failed": 633,
      "total_read": 21044,
      "delivery_rate": 98.4,
      "read_rate": 54.1,
      "avg_delivery_time_ms": 1840
    },
    "time_series": [
      { "period": "2026-05-01", "total_sent": 12345, "total_delivered": 12180, "total_failed": 165, "total_read": 6500, "delivery_rate": 98.66, "read_rate": 53.36, "avg_delivery_time_ms": 1790 },
      { "period": "2026-05-02", "total_sent": 13987, "total_delivered": 13760, "total_failed": 227, "total_read": 7420, "delivery_rate": 98.38, "read_rate": 53.91, "avg_delivery_time_ms": 1875 },
      { "period": "2026-05-03", "total_sent": 13202, "total_delivered": 12961, "total_failed": 241, "total_read": 7124, "delivery_rate": 98.17, "read_rate": 54.95, "avg_delivery_time_ms": 1862 }
    ],
    "group_by": "day"
  },
  "meta": {
    "request_id": "req_xyz789",
    "timestamp": "2026-05-09T00:00:00Z"
  }
}

Query parameters

These filters are shared across the metric and cost endpoints (each endpoint accepts the subset documented below).

Parameter	Description
`start_date` / `end_date`	ISO 8601 timestamps with offset (e.g. `2026-05-01T00:00:00Z`). Both default to the last 30 days when omitted. The range may not exceed 365 days.
`days`	Rolling lookback window, in days, as an alternative to `start_date` / `end_date` (e.g. `7`, `30`, `90`, `365`). Capped at 365.
`channel`	One of `sms`, `whatsapp`, `rcs`, `email`, `viber`, `voice`, `push`, `messenger`, `telegram`.
`status`	(`/messages` only) One of `queued`, `sent`, `delivered`, `failed`, `read`, `undelivered`.
`campaign_id`	(`/messages` only) Restrict the rollup to a single campaign.
`group_by`	(`/messages`, `/costs`) Time bucket for the series: `hour`, `day`, `week`, or `month`. Defaults to `day`.

Endpoints

Endpoint	Method	Description
`/api/v1/analytics/messages`	GET	Aggregate delivery metrics + time-series (the example above).
`/api/v1/analytics/messages/by-channel`	GET	Per-channel delivery breakdown (includes email open / click / complaint counts).
`/api/v1/analytics/messages/by-country`	GET	Per-destination-country delivery breakdown.
`/api/v1/analytics/messages/errors`	GET	Failed messages grouped by error code, with per-code share.
`/api/v1/analytics/costs`	GET	Spend totals, per-currency + per-channel breakdown, and a cost time-series.
`/api/v1/analytics/funnel/{step}/drop-reasons`	GET	Top carrier rejection codes for a funnel step (see below).
`/api/v1/analytics/telegram`	GET	Telegram-specific KPIs (bot-blocked rate, group-vs-DM split, custom-emoji rate).
`/api/v1/analytics/deliverability`	GET	Per-channel deliverability dashboard.
`/api/v1/analytics/journey-paths`	GET	Cross-channel customer-journey path / Sankey (see Customer journey paths).
`/api/v1/analytics/scheduled-reports`	GET / POST	List + schedule recurring report exports (see Scheduled exports).

Cost analytics

curl -G https://api.orbit.devotel.io/api/v1/analytics/costs \
  -H "X-API-Key: dv_live_sk_..." \
  --data-urlencode "channel=sms" \
  --data-urlencode "start_date=2026-05-01T00:00:00Z" \
  --data-urlencode "end_date=2026-05-08T00:00:00Z" \
  --data-urlencode "group_by=day"

Response

{
  "data": {
    "totals": {
      "total_messages": 39534,
      "total_spend": 287.91,
      "avg_cost_per_message": 0.0073,
      "total_segments": 41208,
      "by_currency": [
        { "currency": "USD", "total_messages": 39534, "total_spend": 287.91, "avg_cost_per_message": 0.0073, "total_segments": 41208 }
      ]
    },
    "time_series": [
      { "period": "2026-05-01", "currency": "USD", "total_messages": 12345, "total_spend": 89.92, "avg_cost_per_message": 0.0073, "total_segments": 12870 }
    ],
    "by_channel": [
      { "channel": "sms", "currency": "USD", "total_messages": 39534, "total_spend": 287.91, "avg_cost_per_message": 0.0073, "total_segments": 41208 }
    ],
    "group_by": "day"
  },
  "meta": {
    "request_id": "req_abc123",
    "timestamp": "2026-05-09T00:00:00Z"
  }
}

Field reference

These cost fields count billed messages only — sends that actually incurred a charge. Free-tier, promotional, and zero-priced sends carry no charge and are excluded, so the counts here are lower than the send totals from /analytics/messages.

Field	Definition
`total_messages`	Number of billed messages in the group — sends that incurred a charge. Zero-cost and unpriced sends are excluded, so this is smaller than your total send count.
`total_spend`	Sum of the platform charges for those billed messages, in the group’s `currency`.
`avg_cost_per_message`	`total_spend` divided by `total_messages` — the average charge per billed message, rounded to four decimal places. Because zero-cost sends are excluded from the denominator, this reads higher than dividing `total_spend` by every message you sent. It is a per-message figure, not per-segment. A group with no billed messages returns `0`.
`total_segments`	Total SMS segments across those messages. A long SMS is split into multiple segments, so `total_segments` is at or above `total_messages`; use it to reconcile per-segment carrier charges separately from the per-message average.

Every row is grouped by currency, so a tenant that changed billing currency mid-range gets one row per currency rather than a mixed sum. To reconcile a total against avg_cost_per_message, divide total_spend by total_messages (billed) for the matching currency — not by your overall send count.

Funnel drop-off reasons

Drill into why messages dropped at a funnel step. {step} is one of sent, delivered, read, or queued; the response returns the top carrier rejection codes (default 5, max 20 via limit) enriched with human-readable names and categories from Orbit’s carrier-error dictionary.

curl -G https://api.orbit.devotel.io/api/v1/analytics/funnel/sent/drop-reasons \
  -H "X-API-Key: dv_live_sk_..." \
  --data-urlencode "channel=sms" \
  --data-urlencode "start_date=2026-05-01T00:00:00Z" \
  --data-urlencode "end_date=2026-05-08T00:00:00Z" \
  --data-urlencode "limit=5"

Response

{
  "data": {
    "step": "sent",
    "reasons": [
      {
        "error_code": "30007",
        "name": "Carrier filtered",
        "description": "The carrier blocked the message as suspected spam.",
        "category": "carrier_filtering",
        "count": 318,
        "percentage": 50.2
      },
      {
        "error_code": "30005",
        "name": "Unknown destination",
        "description": "The destination handset is unknown or has been deactivated.",
        "category": "invalid_destination",
        "count": 142,
        "percentage": 22.4
      }
    ]
  },
  "meta": {
    "request_id": "req_def456",
    "timestamp": "2026-05-09T00:00:00Z"
  }
}

Customer journey paths

Reconstruct the ordered touchpoint sequence customers actually walked across channels (chat → voice → email, for example), with per-stage reach and drop-off plus the most-walked full paths. This is the aggregate path view — distinct from the message-delivery funnel above (a single per-message send→deliver funnel) and from the campaign journey builder (the journey a marketer designs).

curl -G https://api.orbit.devotel.io/api/v1/analytics/journey-paths \
  -H "X-API-Key: dv_live_sk_..." \
  --data-urlencode "window=30d" \
  --data-urlencode "max_stages=6" \
  --data-urlencode "entry_channel=chat" \
  --data-urlencode "min_journeys=1"

Query parameters (journey paths)

Parameter	Description
`window`	Look-back window — one of `24h`, `7d`, `30d`, `90d`. Defaults to `30d`.
`entry_channel`	Restrict to journeys whose first touchpoint was on this channel (e.g. `chat`, `voice`, `whatsapp`). Omit to include all entry channels.
`max_stages`	Number of ordered stages to render in the path (defaults to `6`, minimum `2`, maximum `10`).
`min_journeys`	Prune stage-to-stage links carried by fewer than this many journeys (default `1`).

Response

{
  "data": {
    "total_journeys": 1240,
    "total_touchpoints": 5217,
    "max_stages": 6,
    "channels": ["chat", "email", "voice"],
    "nodes": [
      { "stage": 0, "channel": "chat", "journeys": 980 },
      { "stage": 1, "channel": "voice", "journeys": 412 }
    ],
    "links": [
      { "from_stage": 0, "from_channel": "chat", "to_channel": "voice", "journeys": 41 }
    ],
    "stages": [
      {
        "stage": 0,
        "reached": 980,
        "continued": 412,
        "dropped": 568,
        "drop_off_rate": 0.5796,
        "channels": [ { "channel": "chat", "journeys": 980 } ]
      }
    ],
    "top_paths": [
      { "path": ["chat", "voice"], "journeys": 651, "share": 0.5250 }
    ]
  },
  "meta": {
    "request_id": "req_xyz789",
    "timestamp": "2026-07-22T00:00:00Z"
  }
}

Customers are matched across channels by their linked contact record (contact id, else phone, else email), so a chat and a later phone call from the same person count as one journey. Only journeys in the requested window are considered.

Common errors

Code	HTTP	Cause	Fix
`VALIDATION_ERROR`	422	A query parameter failed validation — an unknown `channel` / `status` / `group_by` / `window`, a malformed timestamp, `start_date` not earlier than `end_date`, or a range / `days` value beyond the 365-day lookback cap.	Inspect `error.details.issues` for the offending field; narrow the range or correct the value.
`RATE_LIMITED`	429	Per-tenant rate limit exceeded (60 req/min on analytics endpoints).	Honour `Retry-After` and back off.

Scheduled exports

For ongoing exports, prefer scheduled reports over polling the metric endpoints. POST /api/v1/analytics/scheduled-reports registers a recurring report that Orbit renders and emails to your configured recipients on a daily / weekly cadence; GET /api/v1/analytics/scheduled-reports lists the active schedules. Creating or editing a schedule requires an owner or admin role because the report payload includes per-tenant PII (top contacts, spend, etc.).

Getting Started

Channels

Channels (Voice)

AI Agents

Flows

Numbers & Verify

Concepts

Features

Billing

Webhooks

Security & SSO

Compliance

Legal

Changelog

Cross-channel delivery and engagement analytics

Analytics

Query delivery metrics

Response

Query parameters

Endpoints

Cost analytics

Response

Field reference

Funnel drop-off reasons

Response

Customer journey paths

Query parameters (journey paths)

Response

Common errors

Scheduled exports

​Analytics

​Query delivery metrics

​Response

​Query parameters

​Endpoints

​Cost analytics

​Response

​Field reference

​Funnel drop-off reasons

​Response

​Customer journey paths

​Query parameters (journey paths)

​Response

​Common errors

​Scheduled exports

Analytics

Query delivery metrics

Response

Query parameters

Endpoints

Cost analytics

Response

Field reference

Funnel drop-off reasons

Response

Customer journey paths

Query parameters (journey paths)

Response

Common errors

Scheduled exports